[FFmpeg-cvslog] avfilter/af_surround: avoid divisions with very small numbers

Paul B Mahol git at videolan.org
Mon Apr 22 23:29:06 EEST 2019


ffmpeg | branch: master | Paul B Mahol <onemda at gmail.com> | Mon Apr 22 15:18:58 2019 +0200| [c6c94303d44798f968fa57f282627b1c5e60ce85] | committer: Paul B Mahol

avfilter/af_surround: avoid divisions with very small numbers

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=c6c94303d44798f968fa57f282627b1c5e60ce85
---

 libavfilter/af_surround.c | 27 ++++++++++++++++++---------
 1 file changed, 18 insertions(+), 9 deletions(-)

diff --git a/libavfilter/af_surround.c b/libavfilter/af_surround.c
index 117c2a2d4c..d7690cf12a 100644
--- a/libavfilter/af_surround.c
+++ b/libavfilter/af_surround.c
@@ -910,7 +910,8 @@ static void filter_stereo(AVFilterContext *ctx)
         float l_phase = atan2f(l_im, l_re);
         float r_phase = atan2f(r_im, r_re);
         float phase_dif = fabsf(l_phase - r_phase);
-        float mag_dif = (l_mag - r_mag) / (l_mag + r_mag);
+        float mag_sum = l_mag + r_mag;
+        float mag_dif = mag_sum < 0.000001 ? 0.f : (l_mag - r_mag) / mag_sum;
         float mag_total = hypotf(l_mag, r_mag);
         float x, y;
 
@@ -944,7 +945,8 @@ static void filter_surround(AVFilterContext *ctx)
         float l_phase = atan2f(l_im, l_re);
         float r_phase = atan2f(r_im, r_re);
         float phase_dif = fabsf(l_phase - r_phase);
-        float mag_dif = (l_mag - r_mag) / (l_mag + r_mag);
+        float mag_sum = l_mag + r_mag;
+        float mag_dif = mag_sum < 0.000001 ? 0.f : (l_mag - r_mag) / mag_sum;
         float mag_total = hypotf(l_mag, r_mag);
         float x, y;
 
@@ -977,7 +979,8 @@ static void filter_2_1(AVFilterContext *ctx)
         float l_phase = atan2f(l_im, l_re);
         float r_phase = atan2f(r_im, r_re);
         float phase_dif = fabsf(l_phase - r_phase);
-        float mag_dif = (l_mag - r_mag) / (l_mag + r_mag);
+        float mag_sum = l_mag + r_mag;
+        float mag_dif = mag_sum < 0.000001 ? 0.f : (l_mag - r_mag) / mag_sum;
         float mag_total = hypotf(l_mag, r_mag);
         float x, y;
 
@@ -1018,8 +1021,10 @@ static void filter_5_0_side(AVFilterContext *ctx)
         float sr_phase = atan2f(sr_im, sr_re);
         float phase_difl = fabsf(fl_phase - sl_phase);
         float phase_difr = fabsf(fr_phase - sr_phase);
-        float mag_difl = (fl_mag - sl_mag) / (fl_mag + sl_mag);
-        float mag_difr = (fr_mag - sr_mag) / (fr_mag + sr_mag);
+        float magl_sum = fl_mag + sl_mag;
+        float magr_sum = fr_mag + sr_mag;
+        float mag_difl = magl_sum < 0.000001 ? 0.f : (fl_mag - sl_mag) / magl_sum;
+        float mag_difr = magr_sum < 0.000001 ? 0.f : (fr_mag - sr_mag) / magr_sum;
         float mag_totall = hypotf(fl_mag, sl_mag);
         float mag_totalr = hypotf(fr_mag, sr_mag);
         float bl_phase = atan2f(fl_im + sl_im, fl_re + sl_re);
@@ -1075,8 +1080,10 @@ static void filter_5_1_side(AVFilterContext *ctx)
         float sr_phase = atan2f(sr_im, sr_re);
         float phase_difl = fabsf(fl_phase - sl_phase);
         float phase_difr = fabsf(fr_phase - sr_phase);
-        float mag_difl = (fl_mag - sl_mag) / (fl_mag + sl_mag);
-        float mag_difr = (fr_mag - sr_mag) / (fr_mag + sr_mag);
+        float magl_sum = fl_mag + sl_mag;
+        float magr_sum = fr_mag + sr_mag;
+        float mag_difl = magl_sum < 0.000001 ? 0.f : (fl_mag - sl_mag) / magl_sum;
+        float mag_difr = magr_sum < 0.000001 ? 0.f : (fr_mag - sr_mag) / magr_sum;
         float mag_totall = hypotf(fl_mag, sl_mag);
         float mag_totalr = hypotf(fr_mag, sr_mag);
         float bl_phase = atan2f(fl_im + sl_im, fl_re + sl_re);
@@ -1132,8 +1139,10 @@ static void filter_5_1_back(AVFilterContext *ctx)
         float br_phase = atan2f(br_im, br_re);
         float phase_difl = fabsf(fl_phase - bl_phase);
         float phase_difr = fabsf(fr_phase - br_phase);
-        float mag_difl = (fl_mag - bl_mag) / (fl_mag + bl_mag);
-        float mag_difr = (fr_mag - br_mag) / (fr_mag + br_mag);
+        float magl_sum = fl_mag + bl_mag;
+        float magr_sum = fr_mag + br_mag;
+        float mag_difl = magl_sum < 0.000001 ? 0.f : (fl_mag - bl_mag) / magl_sum;
+        float mag_difr = magr_sum < 0.000001 ? 0.f : (fr_mag - br_mag) / magr_sum;
         float mag_totall = hypotf(fl_mag, bl_mag);
         float mag_totalr = hypotf(fr_mag, br_mag);
         float sl_phase = atan2f(fl_im + bl_im, fl_re + bl_re);



More information about the ffmpeg-cvslog mailing list